© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Jacobian LensNEW

Natural Language

MAY 26 Assistant AxisJAN 26 Circuit TracerAPR 26Steer SAE Evals Exports interp-engineNEW Community Blog Privacy & Terms Contact

QWEN3-32B · 32-RESID-BATCHTOPK-65K · 63807 ｜ Neuronpedia

Home
Qwen3-32B
32-RESID-BATCHTOPK-65K
63807

INDEX

Explanations

second derivative of

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Configuration

adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

No Configuration Found

Embeds

Show PlotsShow ExplanationShow ActivationsShow Test FieldShow SteerShow Link

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ç´¯ç´¯

-0.10

ä¹ĭå®¶

-0.08

ities

-0.08

èĬĥ

-0.08

çļĦèº«å½±

-0.08

Pey

-0.07

çİº

-0.07

aro

-0.07

èĮħ

-0.07

çİĩåħĪ

-0.07

POSITIVE LOGITS

dana

0.10

asive

0.09

 Ø¨Ø°ÙĦÙĥ

0.09

repos

0.09

asu

0.09

errated

0.09

æŀŃ

0.09

bÃ¼

0.08

 Ð¿ÑĢÐ¾Ð²ÐµÐ´ÐµÐ½

0.08

åĮħåŃĲ

0.08

Activations Density 0.378%

No Known Activations